In Exercise 39, we fit a linear regression for the number of monthly domestic visitors to Hawaii (for the years 2002 through 2006) using Time and dummy variables for the months as predictors. The R2 value was 96.6% and a residual plot against Time would show no discernible pattern. The data set for this exercise contains the same data for the period January 2000 through May 2013.

In exercise

a) Fit the linear model from Exercise 39 to this entire time period.
b) Would you use this model? Explain.
c) The impact of what two major events can you see in the plot of residuals against Time?

Transcribed Image Text:

Dependent variable is: Domestic Visitors R squared= 96.5% R squared (adjusted)= 95.8% 12870 with 60 13-47 degrees of freedom Variable Coeff SE (Coeff) t-ratio P-value Intercept 302,921.34 6256.5 48.417 <0.0001 Time Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2,197.59 97.94 22.439 <0.0001 15.741.81 8142.96 1.9333 0.059251 66,348.41 8144.73 8.146 0.0001 29.403.22 8147.67 3.609 0.000744 23.988.62 9151.79 2.943 0.005039 97,126 .83 8157.08 11.907 <0.0001 1 138.669.03 8163.55 16.986 <0.0001 100 ,621.64 8171.18 12.314 <0.0001 -24.572.56 8179.98 3.004 0.004262 9 5.725.25 8189.94 0.699 0.487961 -8,112.14 9201.06 -0.99 0.327651 44.985,86 8213.33 5.477 <0.0001